Atsuto Maki

Received a PhD from KTH (preceded by an MEng from UTokyo) and holds a position as Professor of Computer Science at the School of Electrical Engineering and Computer Science. In the past, he was an associate professor at the Graduate School of Informatics, Kyoto University, and has worked in industry for 10+ years (Toshiba Research Cambridge Lab and Toshiba Corporate R&D) where he developed computer vision systems, e.g. for automatic radar image analysis.

He teaches Machine Learning (DD2421) and serves as Vice Director of the Master’s Programme in Machine Learning(TMAIM). The course has been taken by 5K+ students in the past decade. He also serves as the KTH coordinator for partnerships with the National Institute of Informatics (NII), the RIKEN Center for Advanced Intelligence Project (AIP) and the University of Tokyo.

His main research interests span a broad range of topics in machine learning, deep learning, and computer vision, including motion and object recognition, clustering, subspace analysis, stereopsis, and representation learning, and he holds 30+ patents in the area of image analysis. He is a member of the European Laboratory for Learning and Intelligent Systems (ELLIS) and a board member of Swedish Society for Automated Image Analysis (SSBA).

Current projects include:

• VR (Individual grant)
• WASP AI
• WASP AS
• VINNOVA/NFFP7
• SMaRC project
• Digital Futures

Recent publications (2025–; excluding preprints)

• A. Adelöw, M. Gamba, and A. Maki, Deep Probabilistic Supervision for Image Classification, NeurIPS 2026. (to appear)
• T. Sunada, T. Yamasaki, and A. Maki, Training-Free Determination of Network Width via Neural Tangent Kernel, ICLR 2026. PDF
• X. Zhu, J. Henningsson, P. Mårtensson, L. Hanson, M. Björkman, and A. Maki, Designing Synthetic Active Learning for model refinement in manufacturing parts detection, Journal of Manufacturing Systems, vol. 84, pp. 68–84, 2026. PDF
• M. Nordström, A. Maki, and H. Hult, The Impact Label Noise and Choice of Threshold has on Cross-Entropy and Soft-Dice in Image Segmentation, CVPR 2025. PDF
• Y. Xu, L. Bretzner, T. Wang, and A. Maki, Skor-xG: SKeleton-ORiented Expected Goal Estimation in Soccer, CVPR Workshops 2025. PDF
• X. Zhu, J. Henningsson, D. Li, P. Mårtensson, L. Hanson, M. Björkman, and A. Maki, Domain Randomization for Object Detection in Manufacturing Applications Using Synthetic Data: A Comprehensive Study, ICRA 2025. PDF
• R. Maus and A. Maki, Efficient Object-Centric Learning for Videos, SCIA 2025.

Editted Proceedings (2026)

P. Favaro, Z. Kukelova, A. Maki, A. Rohrbach, K. Schindler, and F. Tombari (eds.), Computer Vision – ECCV 2026, Sweden, Proceedings of the 19th European Conference on Computer Vision, LNCS vols. 17001–17078, Springer, 2026.

Suprevision: currently supervising five PhD students and one postdoc researcher.

News

• Served as Program Co-Chair of ECCV 2026, Sweden. KTH News coverage.
• Visiting Professor, Sapienza University of Rome, October-December 2025.
• Visiting Researcher, The University of Tokyo, April-May 2025.
• Co-organising the 42nd Swedish Symposium on Image Analysis/ 8th Swedish Symposium on Deep Learning, March 12-14, 2025. 
• Organised the WASP International Study Trip to Japan for the WASP Cluster for Manufacturing and Process Control, November 2024. Thank you all the host industries!
• Visiting Researcher, The University of Tokyo, November-December 2023.
• My grant application to Vetenskapsrådet got accepted, called "SeDECI" - Semi-Supervised Deep Learning with Class Imbalanced Data. Thank you everyone who has supported me!